Reproducing the results

1. Preprocess the data

1.1 Yang-Hartwich data.

1.1.1 If using all the genes. cd src/preprocessing python preprocess_placenta_patchify_all_genes.py

1.1.2 If using a selected subset of genes of interest. cd src/preprocessing python preprocess_placenta_patchify_selected_genes.py

1.2 MIBI data.

1.2.1 If using all genes. cd src/preprocessing python preprocess_MIBI_patchify_all_genes.py

1. Train the hypergraph wavelet model. 2.1 If using all the genes. cd src python train.py --data-folder '$ROOT/data/MIBI/patchified_all_genes/'

2.2 If using a selected subset of genes of interest. cd src python train.py --data-folder '$ROOT/data/MIBI/patchified_selected_genes/'

Dependencies

We developed the codebase in a miniconda environment. How we created the conda environment: ```

Optional: Update to libmamba solver.

conda update -n base conda conda install -n base conda-libmamba-solver conda config --set solver libmamba

conda create --name scdata pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia -c anaconda -c conda-forge -y conda activate scdata conda install scikit-learn scikit-image pandas matplotlib seaborn tqdm -c pytorch -c anaconda -c conda-forge -y python -m pip install opencv-python python -m pip install phate python -m pip install meld python -m pip install anndata torch_geometric einops geovoronoi vendi-score python -m pip install numpy==1.26 python -m pip install pyarrow python -m pip install transformers python -m pip install dhg==0.9.4 --no-deps python -m pip install optuna torchmetrics scanpy python -m pip install torch-scatter -f https://data.pyg.org/whl/torch-2.5.0+cu121.html python -m pip install git+https://huggingface.co/ctheodoris/Geneformer

export PROJDIR=/usr/local export PROJLIBDIR=/usr/local/lib/ export PROJ_INCDIR=/usr/local/includes/ brew install proj conda install -c conda-forge pyproj python -m pip install squidpy

```

Debug

If you encounter undefined symbol: cublasLtHSHMatmulAlgoInit, version libcublasLt.so.11, can try the following. export LD_LIBRARY_PATH=/home/cl2482/.conda/envs/scdata/lib/python3.10/site-packages/nvidia/cublas/lib:$LD_LIBRARY_PATH NOTE: replace /home/cl2482/.conda/envs/scdata/lib/python3.10/ with your environment parent directory.